A comparative study of phantom sponge for monocular 3D object detection on edge devices

In this paper, we expand the Phantom Sponge attack to monocular 3D object detection to increase false positives and detection times, thereby impairing the performance of edge devices. We introduce alpha-blending with a Universal Adversarial Patch (UAP) to modulate attack strength. Extensive experime...

Full description

Saved in:
Bibliographic Details
Main Authors: Shaheer Siddiqui, Jinpyung Kim, Jangho Lee
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:ICT Express
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2405959525000220
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850118611910262784
author Shaheer Siddiqui
Jinpyung Kim
Jangho Lee
author_facet Shaheer Siddiqui
Jinpyung Kim
Jangho Lee
author_sort Shaheer Siddiqui
collection DOAJ
description In this paper, we expand the Phantom Sponge attack to monocular 3D object detection to increase false positives and detection times, thereby impairing the performance of edge devices. We introduce alpha-blending with a Universal Adversarial Patch (UAP) to modulate attack strength. Extensive experiments on KITTI, Rope3D, and NuScenes datasets across various hardware GPUs assess the UAP's impact. We also analyze power consumption, temperature, and other metrics on NVIDIA Jetson devices under UAP attacks. Our findings reveal the vulnerability of NMS-dependent models and edge devices to UAP attacks, the generalization of UAP, and the need for robust defenses in real-time applications.
format Article
id doaj-art-bdc5050f706145b08eb0588ff031ac69
institution OA Journals
issn 2405-9595
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series ICT Express
spelling doaj-art-bdc5050f706145b08eb0588ff031ac692025-08-20T02:35:50ZengElsevierICT Express2405-95952025-06-0111356957510.1016/j.icte.2025.02.006A comparative study of phantom sponge for monocular 3D object detection on edge devicesShaheer Siddiqui0Jinpyung Kim1Jangho Lee2Department of Computer Science and Engineering, Incheon National University, Incheon 22012, Republic of KoreaGlobal Bridge Co., Ltd., Incheon 21990, Republic of Korea; Correspondence author.Department of Computer Science and Engineering, Incheon National University, Incheon 22012, Republic of Korea; Correspondence author.In this paper, we expand the Phantom Sponge attack to monocular 3D object detection to increase false positives and detection times, thereby impairing the performance of edge devices. We introduce alpha-blending with a Universal Adversarial Patch (UAP) to modulate attack strength. Extensive experiments on KITTI, Rope3D, and NuScenes datasets across various hardware GPUs assess the UAP's impact. We also analyze power consumption, temperature, and other metrics on NVIDIA Jetson devices under UAP attacks. Our findings reveal the vulnerability of NMS-dependent models and edge devices to UAP attacks, the generalization of UAP, and the need for robust defenses in real-time applications.http://www.sciencedirect.com/science/article/pii/S2405959525000220Object detectionAdversarial attackLatency attackReal-time systemEdge-computing
spellingShingle Shaheer Siddiqui
Jinpyung Kim
Jangho Lee
A comparative study of phantom sponge for monocular 3D object detection on edge devices
ICT Express
Object detection
Adversarial attack
Latency attack
Real-time system
Edge-computing
title A comparative study of phantom sponge for monocular 3D object detection on edge devices
title_full A comparative study of phantom sponge for monocular 3D object detection on edge devices
title_fullStr A comparative study of phantom sponge for monocular 3D object detection on edge devices
title_full_unstemmed A comparative study of phantom sponge for monocular 3D object detection on edge devices
title_short A comparative study of phantom sponge for monocular 3D object detection on edge devices
title_sort comparative study of phantom sponge for monocular 3d object detection on edge devices
topic Object detection
Adversarial attack
Latency attack
Real-time system
Edge-computing
url http://www.sciencedirect.com/science/article/pii/S2405959525000220
work_keys_str_mv AT shaheersiddiqui acomparativestudyofphantomspongeformonocular3dobjectdetectiononedgedevices
AT jinpyungkim acomparativestudyofphantomspongeformonocular3dobjectdetectiononedgedevices
AT jangholee acomparativestudyofphantomspongeformonocular3dobjectdetectiononedgedevices
AT shaheersiddiqui comparativestudyofphantomspongeformonocular3dobjectdetectiononedgedevices
AT jinpyungkim comparativestudyofphantomspongeformonocular3dobjectdetectiononedgedevices
AT jangholee comparativestudyofphantomspongeformonocular3dobjectdetectiononedgedevices